x/sys/unix: standardize Kevent_t Udata as integer, not pointer

[wI2L]

unix.SetKevent only takes mode(eg. Filter field) and flags parameter.

Other fields also have different types depending on the ach/platform, say Udata which is either a *byte or a int32.

Since Go1 compat promise prevent us from changing the SetKevent function, perhaps it could be feasible to add new methods to the Kevent_t type to set the others fields, similar to *Iovec.SetLen for example.

Fflags and Data methods looks straightforward to implement, on the model of SetKevent, since types are integers on all platforms/archs.

However, so far, I am unsure about the type of the Udata parameter for the eponym method. I typically would do kev.Udata = (*byte)(unsafe.Pointer(&foobar)) on my own currently for darwin/amd64, but I guess that exposing unsafe.Pointer as a parameter type is not a good solution.

[bcmills]

CC @tklauser @bradfitz @ianlancetaylor

[ianlancetaylor]

unsafe.Pointer seems to me to be the only reasonable choice for the Udata field. The C type is void * and unsafe.Pointer is the Go equivalent.

[wI2L]

Udata field types:

*byte

• darwin_386
• darwin_amd64
• darwin_arm
• darwin_arm64
• dragonfly_amd64
• openbsd_386
• openbsd_amd64
• freebsd_amd64
• freebsd_386
• freebsd_arm

int32

• netbsd_386
• netbds_arm

int64

• netbsd_amd64

[ianlancetaylor]

Using int32 and int64 is not OK if people store pointers in the field. Conversely, using *byte is not OK if people store non-pointers in the field. Unfortunately, Go does not permit a union of pointer and non-pointer types.

[wI2L]

@ianlancetaylor Any idea why *byte isn't used as type for netbsd ?
According to NetBSD manpage for kevent, Udata is of type intptr_t. A Go uintptr could have been used instead of int32 and int64, right ?

[tklauser]

@wI2L It's an intptr_t on netbsd by definition (see NetBSD/src@537e2d7) which cgo then translates to the respective int{32,64} Go type.

[wI2L]

@tklauser Understood. Can we override this to use a uintptr type instead (assuming that making that change is acceptable) ?

[ianlancetaylor]

There is no perfect choice here, but it seems better for the field to consistently be *byte or unsafe.Pointer on all systems.

[rsc]

Merged #32192 into this one, so we can have one discussion instead of two.